JP2004220216A - San/nas integrated storage device - Google Patents

San/nas integrated storage device Download PDF

Info

Publication number
JP2004220216A
JP2004220216A JP2003005234A JP2003005234A JP2004220216A JP 2004220216 A JP2004220216 A JP 2004220216A JP 2003005234 A JP2003005234 A JP 2003005234A JP 2003005234 A JP2003005234 A JP 2003005234A JP 2004220216 A JP2004220216 A JP 2004220216A
Authority
JP
Japan
Prior art keywords
data
storage device
file server
network
file
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
JP2003005234A
Other languages
Japanese (ja)
Inventor
Akiyoshi Hashimoto
Norio Hirako
Atsushi Tanaka
典夫 平児
顕義 橋本
淳 田中
Original Assignee
Hitachi Ltd
株式会社日立製作所
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hitachi Ltd, 株式会社日立製作所 filed Critical Hitachi Ltd
Priority to JP2003005234A priority Critical patent/JP2004220216A/en
Publication of JP2004220216A publication Critical patent/JP2004220216A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/07Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/16Error detection or correction of the data by redundancy in hardware
    • G06F11/20Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
    • G06F11/202Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where processing functionality is redundant
    • G06F11/2023Failover techniques
    • G06F11/2033Failover techniques switching over of hardware resources
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06Digital input from or digital output to record carriers, e.g. RAID, emulated record carriers, networked record carriers
    • G06F3/0601Dedicated interfaces to storage systems
    • G06F3/0602Dedicated interfaces to storage systems specifically adapted to achieve a particular effect
    • G06F3/0614Improving the reliability of storage systems
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06Digital input from or digital output to record carriers, e.g. RAID, emulated record carriers, networked record carriers
    • G06F3/0601Dedicated interfaces to storage systems
    • G06F3/0628Dedicated interfaces to storage systems making use of a particular technique
    • G06F3/0638Organizing or formatting or addressing of data
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06Digital input from or digital output to record carriers, e.g. RAID, emulated record carriers, networked record carriers
    • G06F3/0601Dedicated interfaces to storage systems
    • G06F3/0628Dedicated interfaces to storage systems making use of a particular technique
    • G06F3/0655Vertical data movement, i.e. input-output transfer; data movement between one or more hosts and one or more storage devices
    • G06F3/0659Command handling arrangements, e.g. command buffers, queues, command scheduling
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06Digital input from or digital output to record carriers, e.g. RAID, emulated record carriers, networked record carriers
    • G06F3/0601Dedicated interfaces to storage systems
    • G06F3/0668Dedicated interfaces to storage systems adopting a particular infrastructure
    • G06F3/067Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/07Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/16Error detection or correction of the data by redundancy in hardware
    • G06F11/20Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
    • G06F11/202Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where processing functionality is redundant
    • G06F11/2038Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where processing functionality is redundant with a single idle spare processing component
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network-specific arrangements or communication protocols supporting networked applications
    • H04L67/10Network-specific arrangements or communication protocols supporting networked applications in which an application is distributed across nodes in the network
    • H04L67/1097Network-specific arrangements or communication protocols supporting networked applications in which an application is distributed across nodes in the network for distributed storage of data in a network, e.g. network file system [NFS], transport mechanisms for storage area networks [SAN] or network attached storage [NAS]

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problems in a storage system directly connected to a network that, when a conventional I/F and protocol is used in transmission of an I/O command issued by a file server to the storage device, performance degradation is caused because data are serially transmitted in a single transfer path, and the unique I/O characteristic of the file system cannot be transferred to an application or a file system and, further, although high-speed and precise fail-over of the file server requires the sharing of a file server-side failure by the storage device side, the conventional I/F has no means for transmitting it. <P>SOLUTION: The file server and a channel adaptor in the storage device are arranged on the same board 112, and the connection path between them is controlled so that a plurality of protocols can be independently operated. A path independent from a path of command and data is disposed between the file server 113 and the channel adaptor 114, and received failure information is stored in a shared memory in the storage device 100 and used for the fail-over. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a device in which a file system and a storage system are integrated and a management method thereof, and more particularly to a technique for processing I / O commands and failure processing from a file system at high speed.
[0002]
[Prior art]
[Patent Document 1]
JP-A-2000-99281
[Patent Document 2]
JP 2002-14878 A
[Non-patent document 1]
US Patent Publication 2002/0116593
2. Description of the Related Art Conventionally, as a storage device used by being connected to a host computer, there is one disclosed in Japanese Patent Application Laid-Open No. 2000-99281. This is a configuration in which a channel adapter connected to a host computer, a disk adapter connected to a disk device, a disk cache, and a control memory are connected by an interconnection network in order to speed up data transfer and control in the storage device. Take.
[0003]
In addition, a file server in the storage device and a host computer are connected by Ethernet so that a user who has only an Ethernet (registered trademark) port can connect, and a fiber channel port block interface is provided between the file server and the channel adapter. Is disclosed in, for example, US Patent Application 2002/0116593.
[0004]
Note that a method of performing communication between a server and storage using a dedicated network, for example, using a fiber channel protocol, is generally referred to as a SAN (Storage Area Network). In addition, a method of directly connecting to a LAN and performing communication using a protocol based on a standard such as a TCP / IP protocol or “Ethernet II” (Ethernet is a registered trademark) “IEEE802.3” or “IEEE802.2” via Ethernet, for example, is used. Generally, it is referred to as NAS (Network Area Storage).
On the other hand, as a computer system having a path for transmitting fault information directly from a file server to a storage device, there is a method disclosed in, for example, JP-A-2002-14878.
[0005]
[Problems to be solved by the invention]
An object of the present invention is to provide a new system in which SAN / NAS is integrated and integrated.
[0006]
FIG. 3 is a system configuration diagram created as a comparative example of the present invention. This system is a SAN / NAS integrated storage system in which a mutual connection network is applied and a file system and a storage system are integrated and integrated.
[0007]
Generally, in order for a user or an application to access data, it is necessary to use a host computer (300), a file server (310), and a storage device (340) as shown in FIG. In this case, the user and the application exist in the host computer (300), and issue a data access request from the host computer (300) via the network or the Ethernet port (305) to the file server (310) that manages the file. The file server (310) manages data (files) recognized by the user and the application, and issues data (block) read / write requests to the disks (160, 161). At this time, the file server (310) converts the file information into block information. The disks (160, 161) store and read data according to commands from the file server (310).
In such a storage device (340), with the increase in the capacity of the disks (160, 161), the improvement in interface performance and processor performance, and the increase in the mounting density of LSIs and boards, the integration of hardware has been promoted. ing. Accordingly, the number of channel adapters and the number of disks increase, and the channel adapters (110, 310), the disk adapters (140, 141), the disk cache (132), and the control memory are used to speed up data transfer and control in the storage device. It is advantageous to adopt a configuration in which the interconnections (131) are connected by an interconnection network (130). As a storage device having an interconnection network, there is a method disclosed in the above-mentioned JP-A-2000-99281.
[0008]
On the other hand, demands for reduction in management costs have led to the aggregation of a plurality of user data into the storage device (340). However, not all users have a block interface represented by Fiber Channel, and many users often have only an IP interface represented by Ethernet. In addition, when connecting to the storage device (340) from a remote area, there is a problem in that the cost increases when the fiber channel is used. Therefore, as shown in FIG. 3, in order to promote the integration of the storage device (340), a file server is built in the storage device (340) so that even a user having only an IP interface such as an Ethernet port (305) can connect. There is an increasing demand for a storage device (340) in which a file server (310) is placed in a storage device (340) and connected to a channel adapter (330) by a block interface such as a fiber channel port (320).
[0009]
As a NAS having such an IP interface, for example, there is a method disclosed in US Patent Application 2002/0116593. The storage device disclosed in this patent application has a file server that controls an IP interface for a user, and is connected to the storage device by a fiber channel via a mutual switch. According to this method, the file server processing and the storage processing are performed separately, so that high performance can be achieved.
[0010]
On the other hand, when the scale of the storage device increases and NAS becomes general, it is necessary to cooperate with failure processing between the file server (315) and the storage device (340) in order to maintain high reliability of the entire system. come. In particular, if the information is not transmitted directly to the storage device (340) at the time of failure on the file server side (315), it takes time to separate the failed part and to fail over to another part, so that reliability and reliability can be increased. The performance decreases. For this reason, a method of transmitting fault information using a different path for fault information separately from a path through which normal data passes is being studied. As such a computer system having a path for directly transmitting fault information from a file server to a storage device, there is a method disclosed in, for example, JP-A-2002-14878. The computer system disclosed in this publication does not use a host bus controlled by a processor on the file server side, and employs a configuration having a separate path for communicating failure information to a storage device. According to this method, a failure on the file server side can be directly known irrespective of the state of the processor, and high reliability of the system can be maintained.
[0011]
However, in the storage device disclosed in JP-A-2000-99281, the interface with the host computer does not include the IP interface. Therefore, a user who does not have a file server cannot connect to this storage device. Therefore, it is necessary to prepare a new file server to connect to this storage device. FIG. 3 shows an example in which a file server (310) is newly provided. However, this comparative example has a problem that the management cost and the installation area increase.
[0012]
In the NAS device disclosed in US Patent Application No. 2002/0116593, since a NAS is provided in the device, a file server and a storage device are connected. However, a connection used to transfer data and commands once. The path is a single Fiber Channel cable, and when the load is high, there is a high possibility that the performance will be degraded. Further, since the connection is made only by the fiber channel cable, when a failure occurs in the file server, the failure information cannot be transmitted to the storage device. Therefore, it is highly likely that much time will be required to complete the failover. In other words, this NAS device has problems of performance degradation and reliability degradation.
[0013]
On the other hand, a computer system disclosed in Japanese Patent Application Laid-Open No. 2002-14878 concerning failure processing has a file server (referred to as an I / O processor in the known example) directly on the disk controller side (referred to as a main processor in the known example). The failure information can be transmitted, but the information is used for failure processing only after the setting information of the file server is changed and transmitted to an external management server through the IP network, so that the load on the IP network is high. In such a case, there is a problem that the other party cannot acquire the failure information accurately or it takes time to transmit the failure information.
Accordingly, an object of the present invention is to provide a system having a file server interface in a storage device and a method thereof.
[0014]
Another object of the present invention is to provide a system and a method for processing commands and data in parallel between a file server and a storage device.
[0015]
Still another object of the present invention is to provide a system and method for transferring failure information of a file server to a storage device separately from normal commands and data, and sharing the information with the entire storage device.
[0016]
Still another object of the present invention is to provide a system and a method for performing failover of a failed file server using failure information shared by storage devices.
[0017]
[Means for Solving the Problems]
In order to solve the above problem, in a storage device according to one embodiment of the present invention, a disk for storing data, a first adapter directly connected to a first network connected to a first host computer, and a disk A first adapter, a first adapter, and a second network directly connected to and coupled to the second adapter, wherein the first adapter is connected to the host via the first network. The information according to the protocol received from the computer is converted into the information according to the second protocol and transferred to the second adapter via the second network. According to the present invention, the host computer and the adapter can be connected by a high-speed interface that does not perform redundant protocol conversion by using an internal bus such as a PCI interface. In particular, in the present invention, the file server and the channel adapter of the storage device are placed on the same substrate as the first adapter, so that they can be connected by a high-speed interface.
[0018]
The first protocol is a so-called Ethernet protocol based on, for example, any of the standards of “Ethernet II”, “IEEE802.3” and “IEEE802.2”, and enables connection with various host computers. The second protocol is, for example, a fiber channel, and can realize a dedicated high-speed channel. According to the present invention, the two can be connected only by the first adapter. Therefore, it is excellent in space factor and maintenance.
[0019]
The file server unit has a server processor that controls the file server unit, a LAN controller that controls data and commands coming from the first network, and a first internal bus that connects the server processor and the LAN controller. For example, the server processor performs data and command conversion processing and sends the data and command to the channel adapter unit according to the second protocol. The channel adapter unit sends the converted data and command to the second network.
[0020]
In this way, by integrating the file server unit and the channel adapter unit, a network data path for transmitting and receiving data and a control data path for transmitting and receiving commands are physically and logically independent between the two. It can be easily provided. That is, a plurality of independent paths are provided between the file server and the storage device so that data and commands can be processed by different paths during I / O processing.
[0021]
Further, the file server unit may include a host controller for transmitting and receiving data and commands between the server processor and the first internal bus, and controlling fault information. Further, the channel adapter unit may include a channel processor that controls the channel adapter unit. It is desirable to provide a management bus between the file server unit and the channel adapter unit for transmitting and receiving fault information without passing through the first internal bus. This management bus directly connects, for example, a server processor or a host controller and a control data controller in a channel adapter unit. Even when the first internal bus cannot be used due to a failure, the failure information of the server processor or the host controller can be transferred to the channel adapter unit. In this way, the failure processing information of the file server can be transferred to the storage device using a path different from the I / O processing. Further, the failure processing information is stored in a memory so that it can be shared by the storage system, and can be referred to from another storage device or a file server that performs a failover.
[0022]
Another aspect of the present invention is the novel first adapter described above, which is, for example, in accordance with one of the standards "Ethernet II", "IEEE 802.3" and "IEEE 802.2" (referred to as "external protocol"). A disk controller that sends and receives commands and data to and from a host computer according to a protocol based on the protocol, and accesses a recording disk according to a protocol other than an external protocol (referred to as an “internal protocol”). A file server unit and a channel adapter unit are provided above.
[0023]
The file server unit controls the file server unit, converts a data and a command from an external protocol to an internal protocol, and performs a conversion process of a command, a LAN controller that controls communication of data and a command by an external protocol, a server processor, and the LAN controller. And a host controller between the server processor and the first internal bus for transmitting and receiving data and commands and controlling fault information.
[0024]
A channel data controller for controlling transmission and reception of data to and from the recording disk; a control data controller for controlling transmission and reception of commands to and from the recording disk; a channel processor for controlling the channel adapter; A second internal bus that connects the data controller and the control data controller, and is connected to the first and second internal buses between the file server unit and the channel adapter unit to transfer data based on an internal protocol; A network data path for transmitting and receiving, and a control data path connected to the first and second internal buses for transmitting and receiving commands based on an internal protocol, independently of each other physically or logically; And a management bus for transmitting and receiving fault information without passing through an internal bus.
[0025]
A storage device according to still another aspect of the present invention includes a plurality of disks that store data, a plurality of control devices that receive an I / O command from a host computer and control the disk according to the I / O command, An interconnection network for coupling a plurality of control means, wherein one of the control devices has a block interface as an interface with a host computer, and another of the control devices has a file interface as an interface with the host computer. It has an interface. Here, the file interface (file system interface) refers to an interface for transmitting and receiving data based on a file name. The block interface (block device interface) refers to an interface for transmitting and receiving data based on a device identifier represented by SCSI, a head block address, the number of blocks, and the like, and is a block indicating a data position in a disk. Access data by specifying an address. As described above, a person who provides a storage device integrating interfaces handling different protocols.
[0026]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, an embodiment of the present invention will be described with reference to FIG.
[0027]
FIG. 1 is a configuration diagram of a first embodiment of the present invention.
[0028]
In FIG. 1, reference numeral 100 denotes a storage device for storing user and application data. The storage device (100) of the present invention has an interface compatible with the host computers (101, 300). The host computer (300) having a fiber channel interface is connected by a fiber channel interface (103) including a plurality of fiber channel ports. A host computer having an IP interface is connected by an Ethernet interface (104, 105) including a plurality of Ethernet ports. Inside the storage device (100), a channel adapter (110) for connecting a fiber channel interface (103) of a host computer (300) and performing data and command processing and an Ethernet interface (104) of the host computer (101) are provided. It has file server boards (112, 115) that are connected via the external network (102) and process data and commands including files. The file server boards (112, 115) have their own OS, process the IP protocol from the host computer (101), and convert the file level requests into block level requests and process the block level requests with the file server unit (113). And a channel adapter section (114). The file server unit and the channel adapter unit (114) are arranged on the same board, and are connected by a plurality of high-speed interfaces as described later. The storage device (100) maintains the consistency of the data in the storage with the disk cache (132) that temporarily stores data and speeds up the reading and writing of data, and saves and shares the state in the device. (133). Further, the storage device (100) has disk adapters (140, 142) for controlling disks (160, 161) for finally storing data. The channel adapter (110), the file server board (112), the disk cache (132), and the control memory (133) The disk adapters (140, 142) have an interface with the interconnection network (130) and are connected to each other. . The interconnection network (130) is composed of a switch or the like, and is a high-speed and highly reliable network as compared with the external network (102). With the above configuration, the storage apparatus of the present invention has a plurality of interfaces in the same apparatus, so that there is no need to prepare an extra file server as shown in FIG. 3, and it is possible to reduce the apparatus cost. Next, details of the file server boards (112, 115) will be described.
[0029]
FIG. 2 is a configuration diagram of a file server board in which a file server unit and a channel adapter unit according to the present invention are integrated.
[0030]
In FIG. 2, reference numeral 200 denotes a file-level command and data transmission / reception to / from a host computer through an external network (102), and a disk cache (132), a control memory (133) and a disk adapter (140) through an interconnection network (130). , 142) and the file server boards (112, 115) for transmitting and receiving block-level commands and data. In FIG. 2, the number of interfaces with the external network (102) and the interconnection network (130) is limited to one for ease of explanation. is there. Reference numeral 201 denotes a file server unit that receives commands and data from the external network (102), performs processing in its own file system, converts block-level data and commands, and sends the converted data to the channel adapter unit (202). send. Reference numeral 210 denotes a server processor, which controls the entire file server unit (201). Reference numeral 211 denotes a host controller, which performs transmission / reception of data, commands, and interrupt signals on the server processor (210) and peripheral memories, an internal bus (214), and control of fault information. A LAN controller 212 controls data and commands coming from the external network (102). An Ethernet 213 connects the external network (102) and the file server board (200) and transmits data and commands conforming to the IP protocol. An internal bus 214 connects the host controller (211), the LAN controller (212), and the channel adapter unit (202), and transmits and receives data and commands. In addition to the internal bus (214), the server processor (210) and the host controller (211) have a management bus (230) as a path for independently transmitting and receiving fault information. The management bus (230) is connected to the control data controller (221) in the channel adapter unit (202). Even if the internal bus (214) cannot be used due to a failure, the server processor (210) and the host controller (211) Is transferred to the channel adapter unit (202). Reference numeral 202 denotes a channel adapter unit which receives block-level data and commands sent from the file server unit (201), and appropriate disk cache (132) and control memory (132) through an interconnection network (130) according to the contents. 133), and sends it to the disk adapter (140, 142). Reference numeral 220 denotes a channel processor which controls the entire channel adapter unit (202). A control data controller 221 controls transmission and reception of data necessary for controlling the entire storage device (100), and controls transmission and reception of failure information from the server processor (210) and the host controller (211). Reference numeral 222 denotes an interconnection network (130) for the control memory (133), which transfers control data from the control data controller to the control memory (133), disk adapters (140, 142), other channel adapters (110), and a file server. Transfer to the board (113). A network data controller 223 controls transmission and reception of data of a user or an application. Reference numeral 224 denotes an interconnection network (130) for transferring user and application data, and transfers data from the network data controller (223) to the disk cache (132) and the disk adapters (140, 142). A control data path 231 transmits and receives commands and information necessary for controlling the entire storage device (100) between the file server unit (201) and the channel adapter unit (202). Reference numeral 232 denotes a network data path for transmitting and receiving user or application data, parameters necessary for data transfer, and the like between the file server unit (201) and the channel adapter unit (202). Reference numeral 250 denotes a management network, which is used to transfer fault information and configuration information collected by the channel processor (220) to the management processor (255). The management processor (255) determines failure processing, configuration change, and the like based on the received information, and issues an instruction to the storage device (100) if necessary. Further, the management processor (255) can integrally display information of each board on the management display (260). For example, a management table such as 261 can be displayed, and the status of the file server unit (266) and the channel adapter unit (267) can be displayed for each board # (265). Reference numeral 240 denotes a power supply for the file server board (200), which supplies power to the file server unit (201) and the channel adapter unit (202). With the configuration described above, the interface between the file server and the storage device can be integrated on the same board, and the installation area of the device can be reduced. In addition, the file server and the storage device can be managed on the same management processor (255) and management display (260), thereby reducing management costs. Further, with regard to data and command transfer between the file server unit (201) and the channel adapter unit (202), by providing two types of independent paths of a control data path (231) and a network data path (232), commands and parameters are provided. By separating and processing relatively short data such as user data and relatively long data such as user and application data, data transfer performance can be improved. In addition, commands and the like can be transferred collectively, and processing efficiency can be increased. Further, by separating the command from the data, the format of the command can be changed from the standard specification, so that information and parameters specific to the device can be directly transmitted to the channel adapter, and selection of resident data in the disk cache (132) can be performed. It becomes possible. These two types of independent paths are composed of two physically separated paths, and a physically single path as shown in the hardware configuration diagram of the file server board (112) described below. However, it is possible to implement using either of the methods of controlling so that there are two paths logically, but it is obvious that they are within the scope of the present patent. FIG. 2 shows an example in which two types of processors, a server processor (210) and a channel processor (220), are operating on the file server board (112). However, other than this embodiment, only the server processor (210) is used. It is clear that a scheme for controlling all the components on the file server board (112) can also be implemented, which is also within the scope of the present patent.
[0031]
FIG. 4 is a hardware configuration diagram of the file server board according to the first embodiment of the present invention.
[0032]
In FIG. 4, a host bus (413) is connected between a server processor (210) in a file server unit (400) and a host controller (211). The host controller (211) is connected with a management bus controller (512) for controlling the management bus (555), a local memory (410) for storing programs and data of the server processor (210), and an internal bus (214, 225). And an internal bus controller (511) for controlling the operation. The internal bus controller (411) controls a path connecting the file server unit (400) and the channel adapter unit (460), and controls the internal buses (214, 225) so that there are logically a plurality of paths. That is, the control data path (231) and the network data path (232) in FIG. 3 can be logically configured. The channel processor (220) in the channel adapter section (460) is connected to the local bus (471) via the local bus controller (470). Although not shown in FIG. 4, a memory and the like necessary for the channel processor (220) are connected to the local bus. The local bus controller (470) is connected to the management network (250) and transmits necessary management information and the like to the management processor (225). The control data controller (221) is connected to the local bus (471). An internal bus (255) is connected to the control data controller (221), and normal control data is transmitted and received. Further, it is connected to a control memory (131) and another channel processor through an interconnection network (222), and transmits and receives control information and shared information. Further, interrupt signals (456, 457) from the server processor (210) and the management bus controller (225) are directly input. Also, a management bus (455) is connected so that information of the server processor can be obtained even when the internal buses (214, 225) are not operating due to a failure or the like. Inside the control data controller (221), a normal register (480) for performing normal control, an error register (481) for storing error information, and a doorbell for controlling communication with the file server unit (400) There are a register (482) and a communication memory (483). A network data controller (223) is connected to the internal bus (225) in addition to the control data controller (221), and transmits and receives user and application data. Also, the control data controller (221) and the management bus controller (225) are connected by a power control means (458) so that the channel processor (220) can control the start, stop, and restart of the file server unit (400). Have been. In the present embodiment, it is assumed that the power of the file server unit can be controlled by the power control unit (458). The configuration of FIG. 4 is merely one example of implementation, and it is obvious that other configurations are also within the scope of the present invention. Next, an example of the structure of a command transmitted and received using the control data path (231) in the solid file server board (200) will be described.
[0033]
FIG. 5 is a diagram illustrating a format of a command data block according to the first embodiment of the present invention.
[0034]
In FIG. 5, reference numeral 500 denotes a command data block transmitted from the file server unit (400) to the channel adapter unit (202). A command type 501 distinguishes between a normal command and a special command. Reference numerals 502, 510, 535, and 540 are reserved areas and have no particular meaning. Reference numeral 503 denotes a TAG type, which instructs a control method when stored in a queue. Reference numeral 504 denotes a server processor #, which specifies the server processor (210) that has issued this command. Reference numeral 505 denotes an IID that distinguishes access to a user area or a system area. Reference numeral 506 denotes a LUN, which is the number of a logical unit to which the server processor (210) accesses. Reference numeral 511 denotes a TAG #, which indicates an identification number inside the queue. Reference numeral 515 denotes a CDB, in which a command conforming to a format used in normal SCSI is entered. An operation code 516 in the CDB distinguishes commands such as read and write. Reference numeral 520 denotes an initiator port number for distinguishing a source port when a command is transmitted to the outside. Reference numeral 525 denotes an external device WWN, which indicates a destination WWN when a command is transmitted to the outside. An external device LUN 530 indicates a transmission destination LUN when a command is transmitted to the outside. Numeral 541 indicates the number of address entries, and indicates the number of parameters instructing the storage device (100) following the address entry. Reference numerals 545 and 560 denote LBA1 and LBA2, which indicate relative addresses within the LU. Reference numerals 550 and 565 indicate physical addresses in the local memory (410) of the server processor (210), and are storage locations of target data. Reference numerals 555 and 570 denote storage device instruction parameters, which place information specific to the file server unit here and transmit it to the channel adapter. Reference numerals 551 and 571 indicate a buffer size or transfer data length of data corresponding to each parameter. An example of the storage device instruction parameter is a data resident instruction to the disk cache (132). By using this, it is possible to grasp the user, the usage frequency of the application data, and the like that cannot be known only by the storage device (100), and it is possible to improve the performance of the storage device (100). How the command data block described above is transmitted and received among the server processor (210), the control data controller (221), the channel processor (220), and the disk cache (132) will be described below.
[0035]
FIG. 6 is a diagram showing a flowchart at the time of data access according to the first embodiment of the present invention.
[0036]
The server processor (210) and the channel processor (220) come next to a command queue (CmdQueue) (602, 607), a pointer (CmdHead) (601, 606) indicating the head of the queue, respectively, in order to efficiently process the command. It has a pointer (CmdNext) (603, 608) indicating the position where the command should be stored. The command queues (602, 607) are ring buffers. In order to manage the status after completion of the command, a status queue (StQueue) (632, 637), a pointer (StHead) (631, 636) indicating the head of the queue, and a pointer indicating a position where the next command should be stored (StNext) (633, 638). The status queue (StQueue) (632, 637) is also a ring buffer. At the time of command and data transfer, the contents of the command queue (CmdQueue) (602, 607) are checked at any time between the server processor (210) and the channel processor (220). In both cases, the contents of CmdHead (601, 606) and CmdNext (603, 608) need to match. Therefore, the CmdHead (606), CmdQueue (602), and CmdNext (603) are transferred to the server processor (210), the channel processor (220), and the channel processor (220) through the control data controller (221), and the consistency is checked. Take. Similar processing is necessary for the status queue (StQueue) (632, 637), and the server processor (210) and the channel processor (220) check the contents of the status queue (StQueue) (632, 637) as needed. . In order to perform command and data processing, the process (600) of the server processor (210) performs the following three steps.
(1) After confirming that there is a free space in the command queue (CmdQueue) (602), the command is stored.
(2) Update the queue pointer and control CmdNext (603) so that the ring buffer does not overflow.
(3) Start polling and check for status queue updates.
The process (605) of the channel processor (220) performs the following five steps for this process.
(4) Check the update of the command queue (CmdQueue) (607) by polling.
(5) The contents of the command queue CmdQueue (607) are received by DMA transfer.
(6) Check for update of the status queue (StQueue) (637) by polling.
(7) Update the pointer (CmdHead) (606) indicating the head of the queue.
(8) Process each command.
[0037]
Next, in the process (610) of the server processor (210), the parameters (611) necessary for command processing are transferred to the channel processor (220) through the control data controller (221) by DMA transfer. At the same time, the process (620) of the server processor (210) transfers the data by DMA transfer to the disk cache (132) through the network data controller (223). The process (610) and the process (620) can be processed by independent paths as shown in FIGS. 3 and 4, so that the processes can proceed in parallel. After the data transfer processing is completed, the following two steps of processing are performed in the process (635) of the channel processor (220).
(9) Store the execution result in the status queue (StQueue) (637).
(10) Set the doorbell register (582) and generate an interrupt.
For this process, the following three steps are performed in the process (630) of the server processor (210).
(11) Update of status queue (StQueue) (632) is detected by interruption or polling
(12) Status retrieval
(13) Update of the queue pointer (StNext) (633)
As shown in the above flowchart, it is possible to have a queue for each command and status, and each can be processed using an independent path, so that data access performance can be improved.
[0038]
FIG. 7 is a diagram illustrating a flowchart when a server processor failure occurs in the first embodiment of the present invention.
[0039]
In FIG. 7, reference numeral 700 indicates that a failure has occurred in the server processor (210) and it is necessary to perform exception processing. Here, it cannot be expected that the internal buses (214, 225) and the internal bus controller (411) through which normal commands and data are transmitted and received operate normally. The fault information is processed by the exception handler in the process (810), and an error report is recorded in the error register (482) of the control data controller (221) by an interrupt signal in the process (715). Thereafter, the server processor (210) waits for a reset command. Next, the control data controller (221) informs the channel processor (220) in the form of an error interrupt that the server processor (210) has failed in the process (715). In the process (720) of the channel processor (220), the error processing function is activated in response to the error interrupt. Thereafter, the channel processor (220) sends a reset command to the normal register (483) of the control data controller (221) in a process (715) to initialize the server processor (210). The control data controller (221) then sends a reset to the server processor (210) by an interrupt signal. The process (710) of the server processor (210) that has received the reset restarts in the next process (725). The process (725) collects the cause of failure in the server processor (210) and stores it in the communication memory (481) of the control data controller (221) in the process (730). However, since this failure factor is normally transferred through a path, there is no guarantee that the value is correct. After storing the information, the process (725) transmits a save completion report to the channel processor (220) via the doorbell register (482). Thereafter, the channel processor (220) reads out the fault information via the management bus (455) in the process (750) and the process (715) in order to secure correct fault information. Specifically, a chip set register read is performed. At the same time, the cause of failure stored in the communication memory (481) of the control data controller (221) is read out in the process (745). After reading this information, the channel processor (220) reports the information to the management processor (255) and receives a processing instruction. Next, failure information is transferred to the control memory (131), and a failure processing request is issued to another board. Next, it is checked in the process (755) whether an error has occurred in the internal bus (214, 225). If there is no error, the channel processor (220) issues a memory dump instruction to the network data controller (223) in a process (770). The network data controller (223) dumps information and the like in the local memory (410) by DMA transfer in the process (765). If an error has occurred in the internal bus (214, 225), no memory dump is performed. Next, the channel processor (220) issues a forced stop command to the control data controller (221) in the process (790) in response to an externally requested faulty file server board stop request. The control data controller (221) has a function of forcibly stopping the server processor (210) by the power supply control means (458). Upon receiving the forcible stop instruction, the server processor (210) is forcibly closed. In the above process, since a path for transmitting fault information is provided in addition to the normal bus, even if a fault occurs in the bus, the fault information can be transmitted to the channel processor (220), and the fault processing can be highly reliable. Can be achieved. FIG. 7 shows a method of communicating failure information in a single file server board (112), and a method for performing failover between a plurality of file server boards (112) using the method will be described below.
[0040]
FIG. 8 is a diagram illustrating a flowchart of a failover when a server processor fails according to the first embodiment of this invention. Reference numeral 800 indicates that a failure has occurred in the server processor (210) and exception processing needs to be performed as in FIG. The channel processor (220) that has detected the failure performs the processing when a failure occurs in the server processor described with reference to FIG. The failure information and the failure processing request are written into the control memory (131). Thereafter, the file server board (115) at the failover destination detects a failure process in the process (810). Thereafter, the file server board (115) issues a stop request to the file server board (112) where the failure has occurred. Thereafter, the process (815) waits until a stop completion notification is received. Upon receiving the file server board stop request from the file server board (115), the file server board (112) stops the server processor as described with reference to FIG. Further, after discarding the remaining I / O processing in the process (820), the stop of the file server board (112) is written into the control memory (131), and the channel processor is stopped. The control memory (131) having received the stop of the file server board (112) updates the configuration information of the storage device (100) in the process (825) and notifies the failover to the file server board (115) of the stop. . Upon receiving the stop notification, the failover destination file server board (115) takes over the file server in the process (930), notifies the management processor (255) of the result, and completes the failover.
[0041]
When the SAN / NAS integrated storage device disclosed in this patent is used according to the above-described embodiment, in a normal I / O process, a command queue (CmdQueue) (CmdQueue) for each server processor (210) and channel processor (220) is used. 602, 607) and the status queue (StQueue) (632, 637), it is possible to advance command processing and status processing asynchronously, and it is also possible to process commands and data using independent paths. As a result, I / O processing performance can be improved. Further, by using independent paths for commands and data, and by extending the commands, it becomes possible to convey the I / O characteristics unique to the file server to the storage device, and to improve the hit rate of the disk cache (132). The storage device (100) can be optimized. Further, since the file server and the channel adapter of the disk controller are provided on the same board and can be managed by the same management processor (255) and management display (260), the installation area can be reduced and the management cost can be reduced. it can. In the failure processing, failure information is transmitted from the management bus (455) and written to the control memory (131) shared by the storage device (100), so that it can be transmitted to other file server boards regardless of the failure of the bus. As a result, failure information can be quickly transmitted, thereby improving the efficiency of failover and improving reliability.
[0042]
(Example 2)
FIG. 9 is a diagram illustrating a flowchart of a failover when a server processor fails according to the second embodiment of this invention.
In FIG. 8, the file server board (112) of the failure origin has independently notified the occurrence of the failure to the file server board (115) of the failover destination. However, the failure may not be notified due to the failure. Can't start. In order to avoid such a case, a method in which the file server board (115) at the failover destination actively monitors another file server board (112) is considered. If normal, the file server board (112) transmits a heartbeat to the failover destination file server board (115) at regular time intervals. If the heartbeat is not notified within a certain period of time after the occurrence of a failure as in the process (900), the failover destination file server board (115) determines that the failure has occurred, and performs the processing in the process (905). Start fault handling. After the failure source file server board (112) writes the failure information to the control memory (131), the failover destination file server board (115) reads the configuration information to the control memory (131), and a failure occurs. Confirm that this is the case and determine the details of the failure processing. Thereafter, in the process (905), the file server board (115) issues a stop request to the file server board (112) where the failure has occurred. Thereafter, as in FIG. 8, the process enters a wait state of the process (920) until a notification of the completion of the stop comes. Upon receiving the file server board stop request from the file server board (115), the file server board (112) stops the server processor as described with reference to FIG. Further, after discarding the remaining I / O processing in the process (925), the stop of the file server board (112) is written to the control memory (131), and the channel processor is stopped. The control memory (131) having received the stop of the file server board (112) updates the configuration information of the storage device (100) in the process (930) and notifies the stop to the file server board (115) of the failover destination. . Upon receiving the stop notification, the failover destination file server board (115) takes over the file server in the process (935), notifies the management processor (255) of the result, and completes the failover.
According to the embodiment described above, when the disclosed SAN / NAS integrated storage device is used, even when the failure source cannot be transmitted from the file server board (112) of the failure source, the file server board of the failover destination can be obtained at an early stage. (115) can detect and confirm the occurrence of a failure, so that the efficiency of failover can be increased and the reliability can be improved.
[0043]
【The invention's effect】
According to the present invention, since I / O commands and data can be transmitted and received independently between the file server and the disk controller, there is an effect that I / O processing performance can be improved. Further, since the I / O characteristic information of the application and the file system can be given to the disk controller, there is an effect that the I / O processing can be optimized. Further, since the file server and the channel adapter of the disk controller are mounted on the same board and managed by the unified management processor and management processor, there is an effect that the installation area of the device can be reduced and the management cost can be reduced. In addition, since the failure information of the file server can be shared by an independent path, there is an effect that the efficiency of the failover can be increased and the reliability can be increased.
[Brief description of the drawings]
FIG. 1 is a configuration diagram of a SAN / NAS integrated storage system of the present invention.
FIG. 2 is a configuration diagram of a file server board in which a file server unit and a channel adapter unit according to the present invention are integrated.
FIG. 3 is a configuration diagram of a SAN / NAS integrated storage system of a comparative example.
FIG. 4 is a hardware configuration diagram of a file server board according to the first embodiment.
FIG. 5 is a diagram showing a format of a command data block according to the first embodiment.
FIG. 6 is a view showing a flowchart at the time of data access according to the first embodiment;
FIG. 7 is a diagram illustrating a flowchart when a server processor failure occurs according to the first embodiment.
FIG. 8 is a diagram illustrating a flowchart of failover when a server processor fails according to the first embodiment.
FIG. 9 is a diagram illustrating a flowchart of a failover when a server processor fails according to the second embodiment.
[Explanation of symbols]
100, 101, 150, 151: Fiber Channel interface, 102: Ethernet interface, 110: Channel adapter, 112: File server board, 113: File server unit, 114: Channel adapter unit, 130: Interconnection network, 131: Disk cache 132, control memory; 140, 142: disk adapter; 160, 161: drive.

Claims (25)

  1. A plurality of disks for storing data,
    A plurality of control devices for receiving an I / O command from a host computer and controlling the disk according to the I / O command;
    Having an interconnection network for coupling the plurality of control means,
    One of the control devices has a block interface as an interface with a host computer,
    Another one of the control devices has a file interface as an interface with a host computer, and is a SAN / NAS integrated storage device.
  2. The control device having the above block interface
    A channel adapter for communicating with the host computer,
    2. The SAN / NAS integrated storage device according to claim 1, further comprising a disk adapter for controlling the disk.
  3. The control device with the above file interface
    A file server unit that communicates with the host computer;
    A file system for converting file-level commands and data received from the host computer to a block interface,
    A channel adapter unit having an interface with the interconnection network,
    2. The SAN / NAS integrated storage device according to claim 1, further comprising a disk adapter for controlling the disk.
  4. 2. The SAN / NAS integrated storage device according to claim 1, further comprising a control memory connected to the interconnection network and shared by the plurality of control devices.
  5. The control device with the above file interface
    A file server unit that communicates with the host computer;
    A file system for converting file-level commands and data received from the host computer to a block interface,
    A channel adapter unit having the interconnection network and the interface,
    The SAN / NAS integrated storage device according to claim 1, wherein the storage device is mounted on a single board.
  6. The control device with the above file interface
    A file server unit that communicates with the host computer;
    A file system for converting file-level commands and data received from the host computer to a block interface,
    A channel adapter unit for receiving block-level commands and data issued from the file system,
    2. The SAN / NAS integrated storage device according to claim 1, further comprising a disk adapter for controlling the disk.
  7. The control device with the above file interface
    A file server unit that communicates with the host computer;
    2. The SAN / NAS integrated type according to claim 1, wherein a plurality of communication paths are provided between the channel adapter units for receiving block-level input / output commands and input / output data, and commands and data are transmitted / received through separate paths. Storage device.
  8. The control device with the above file interface
    A plurality of independent virtual paths exist between a physical communication path between a file server unit for communicating with the host computer and a channel adapter unit for receiving block-level input / output commands and input / output data. 2. The SAN / NAS integrated storage apparatus according to claim 1, wherein the command and data are transmitted and received through separate paths.
  9. The control device with the above file interface
    A plurality of independent virtual paths exist between a physical communication path between a file server unit for communicating with the host computer and a channel adapter unit for receiving block-level input / output commands and input / output data. 2. The SAN / NAS integrated storage device according to claim 1, wherein the control is performed such that information unique to the file system is inserted into the command, and the command and the data are transmitted and received through separate paths.
  10. The control device with the above file interface
    It has a communication path for failure information between the file server unit that communicates with the host computer and the channel adapter unit that receives block-level input / output commands and input / output data, independent of the command and data communication paths. The SAN / NAS integrated storage device according to claim 1, wherein the failure information is transmitted to a control memory.
  11. A disk for storing data,
    A first adapter directly connected to a first network connected to the first host computer;
    A second adapter directly connected to the disk,
    A second network that is directly connected to the first adapter and the second adapter and connects them;
    The first adapter converts information according to a first protocol received from a host computer via the first network into information according to a second protocol, and converts the information to the second adapter via the second network. Storage device to transfer to.
  12. A third adapter connected to the second host computer via a third network;
    A second network that couples the disk, a first adapter, a second adapter, and a third adapter;
    12. The storage device according to claim 11, wherein the communication protocol on the first network is a first protocol, and the communication protocols on the second and third networks are a second protocol.
  13. 13. The storage device according to claim 12, wherein the first protocol is a protocol based on any of the standards of "Ethernet II", "IEEE802.3", and "IEEE802.2".
  14. The first adapter is
    A file server for transmitting and receiving file-level commands and data to and from the host computer,
    A channel adapter unit for transmitting and receiving block-level commands and data to and from the second adapter;
    Has,
    13. The storage device according to claim 12, wherein they are configured on a single board or in a single housing.
  15. The above file server unit,
    A server processor for controlling the file server unit;
    A LAN controller for controlling data and commands coming from the first network;
    A first internal bus connecting the server processor and the LAN controller;
    And performs data and command conversion processing and sends it to the channel adapter section,
    The channel adapter section is
    15. The storage device according to claim 14, wherein the data and the command subjected to the conversion processing are sent to the second network.
  16. Between the file server section and the channel adapter section,
    A network data path for transmitting and receiving the data,
    A control data path for transmitting and receiving the command,
    16. The storage device according to claim 15, wherein the storage device is physically or logically independent, and the network data path and the control data path are connected to the first internal bus.
  17. The channel adapter section is
    A network data controller that controls transmission and reception of the data to and from the second network;
    A control data controller for controlling transmission and reception of the command to and from the second network;
    A second internal bus connecting the network data controller and the control data controller;
    17. The storage device according to claim 16, wherein the second internal bus is connected to the network data path and the control data path.
  18. The above file server unit,
    A host controller that controls the data, command transmission / reception, and fault information between the server processor and the first internal bus;
    The channel adapter section is
    A channel processor for controlling the channel adapter unit;
    A management bus for connecting at least one of the server processor and the host controller and the channel adapter unit between the file server unit and the channel adapter unit and transmitting / receiving fault information without passing through the first internal bus 18. The storage device according to claim 17, comprising:
  19. The management bus is connected to a control data controller in the channel adapter unit, and transfers failure information of the server processor or the host controller to the channel adapter unit even when the first internal bus cannot be used due to a failure. 19. The storage device according to claim 18, wherein
  20. A management network connected to the channel processor;
    A management processor connected to the management network,
    The channel processor is connected to a control data controller, collects the fault information, transfers the fault information to the management processor,
    20. The storage device according to claim 19, wherein the management processor instructs the channel processor based on the received information.
  21. A control memory connected to the second network,
    The management processor stops the operation of the first adapter having the failed file interface based on the failure information in the control memory, and causes the first adapter having the other normal file interface to operate the first adapter having the normal file interface. 22. The storage device according to claim 21, wherein the operation is substituted.
  22. The management processor compares the fault information obtained from the channel processor with the fault information in the control memory, specifies a control unit having a file interface in which a fault has occurred, and selects a fault handling method. 22. The storage device according to claim 21, wherein:
  23. 22. The storage device according to claim 21, wherein the first adapter transmits a status of the file server unit and the channel adapter unit to the management processor, and receives a management command based on the status.
  24. 21. The storage device according to claim 20, wherein the management processor is connected to a management display, and displays information based on the failure information on the management display.
  25. It transmits and receives commands and data to and from the host computer by a protocol based on any of the standards of “Ethernet II”, “IEEE802.3” and “IEEE802.2” (referred to as “external protocol”). Protocol), the disc controller accesses the recording disc,
    The control device has a file server unit and a channel adapter unit on a single board or in a housing,
    The above file server unit,
    A server processor that controls the file server unit and converts data and commands from the external protocol to the internal protocol,
    A LAN controller for controlling communication of data and commands by the external protocol,
    A first internal bus connecting the server processor and the LAN controller;
    A host controller that is located between the server processor and the first internal bus and controls transmission and reception of the data and commands, and control of fault information;
    The channel adapter section is
    A network data controller that controls transmission and reception of the data to and from the recording disk;
    A control data controller for controlling transmission and reception of the command to and from the recording disk;
    A channel processor that controls the channel adapter unit;
    A second internal bus connecting the network data controller and the control data controller,
    Between the file server section and the channel adapter section,
    A network data path connected to the first and second internal buses for transmitting and receiving data based on the internal protocol;
    A control data path that is connected to the first and second internal buses and transmits and receives commands based on an internal protocol, independently of physically or logically;
    A disk control device having a management bus for transmitting and receiving fault information without passing through the first and second internal buses.
JP2003005234A 2003-01-14 2003-01-14 San/nas integrated storage device Pending JP2004220216A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2003005234A JP2004220216A (en) 2003-01-14 2003-01-14 San/nas integrated storage device

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
JP2003005234A JP2004220216A (en) 2003-01-14 2003-01-14 San/nas integrated storage device
EP03019214A EP1439453A3 (en) 2003-01-14 2003-08-25 SAN/NAS integrated storage system
US10/652,653 US7185143B2 (en) 2003-01-14 2003-08-28 SAN/NAS integrated storage system
US11/701,144 US7697312B2 (en) 2003-01-14 2007-01-31 SAN/NAS integrated storage system

Publications (1)

Publication Number Publication Date
JP2004220216A true JP2004220216A (en) 2004-08-05

Family

ID=32588493

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2003005234A Pending JP2004220216A (en) 2003-01-14 2003-01-14 San/nas integrated storage device

Country Status (3)

Country Link
US (2) US7185143B2 (en)
EP (1) EP1439453A3 (en)
JP (1) JP2004220216A (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2854020A1 (en) 2013-09-27 2015-04-01 Fujitsu Limited Storage control apparatus, storage control method, and storage control program
US9063853B2 (en) 2011-05-31 2015-06-23 Fujitsu Limited Storage device, storage system, and method for controlling storage device
WO2016020979A1 (en) * 2014-08-05 2016-02-11 株式会社日立製作所 Computer system and intermediary device
WO2016075765A1 (en) * 2014-11-12 2016-05-19 株式会社日立製作所 Computer system and control method therefor
US9928185B2 (en) 2015-01-30 2018-03-27 Fujitsu Limited Information processing apparatus and computer-readable recording medium having program recorded therein
JP2018524751A (en) * 2015-07-22 2018-08-30 華為技術有限公司Huawei Technologies Co.,Ltd. Computer device and method for reading / writing data by computer device

Families Citing this family (43)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3964212B2 (en) * 2002-01-16 2007-08-22 株式会社日立製作所 Storage system
JP2004220216A (en) * 2003-01-14 2004-08-05 Hitachi Ltd San/nas integrated storage device
JP4274523B2 (en) * 2003-01-24 2009-06-10 株式会社日立製作所 Storage device system and start method of storage device system
US7565566B2 (en) * 2003-04-23 2009-07-21 Dot Hill Systems Corporation Network storage appliance with an integrated switch
US7627780B2 (en) * 2003-04-23 2009-12-01 Dot Hill Systems Corporation Apparatus and method for deterministically performing active-active failover of redundant servers in a network storage appliance
WO2004095304A1 (en) 2003-04-23 2004-11-04 Dot Hill Systems Corporation Network storage appliance with integrated redundant servers and storage controllers
US7225274B2 (en) * 2003-05-23 2007-05-29 Applied Micro Circuits Corporation Method and apparatus for transferring data across a protocol bridge
JP4437650B2 (en) * 2003-08-25 2010-03-24 株式会社日立製作所 Storage system
US7693960B1 (en) * 2003-10-22 2010-04-06 Sprint Communications Company L.P. Asynchronous data storage system with geographic diversity
JP4257783B2 (en) * 2003-10-23 2009-04-22 株式会社日立製作所 Logically partitionable storage device and storage device system
JP2005135065A (en) 2003-10-29 2005-05-26 Hitachi Ltd Storage device controller and control method for storage device controller
JP4477365B2 (en) * 2004-01-29 2010-06-09 株式会社日立製作所 Storage device having a plurality of interfaces and control method of the storage device
JP4141391B2 (en) 2004-02-05 2008-08-27 株式会社日立製作所 Storage subsystem
JP4521206B2 (en) * 2004-03-01 2010-08-11 株式会社日立製作所 Network storage system, command controller, and command control method in network storage system
JP2005250938A (en) * 2004-03-05 2005-09-15 Hitachi Ltd Storage control system and method
JP2005267008A (en) * 2004-03-17 2005-09-29 Hitachi Ltd Method and system for storage management
JP2005267111A (en) * 2004-03-17 2005-09-29 Hitachi Ltd Storage control system and method for controlling storage control system
JP2005275829A (en) * 2004-03-25 2005-10-06 Hitachi Ltd Storage system
JP4575028B2 (en) * 2004-05-27 2010-11-04 株式会社日立製作所 Disk array device and control method thereof
US7240154B2 (en) * 2004-06-28 2007-07-03 Emc Corporation Low cost flexible network accessed storage architecture
FR2880444B1 (en) * 2005-01-06 2007-03-09 Gemplus Sa Data storage device
CN100354833C (en) * 2005-01-14 2007-12-12 英业达股份有限公司 Protection method for SAN system in abnormal condition
JP4733399B2 (en) * 2005-01-28 2011-07-27 株式会社日立製作所 Computer system, computer, storage device and management terminal
US7519851B2 (en) * 2005-02-08 2009-04-14 Hitachi, Ltd. Apparatus for replicating volumes between heterogenous storage systems
JP2006285464A (en) * 2005-03-31 2006-10-19 Hitachi Ltd Computer system, storage, and device control method
US7676702B2 (en) * 2006-08-14 2010-03-09 International Business Machines Corporation Preemptive data protection for copy services in storage systems and applications
US20080065704A1 (en) * 2006-09-12 2008-03-13 Microsoft Corporation Data and replica placement using r-out-of-k hash functions
US20080168161A1 (en) * 2007-01-10 2008-07-10 International Business Machines Corporation Systems and methods for managing faults within a high speed network employing wide ports
US20080168302A1 (en) * 2007-01-10 2008-07-10 International Business Machines Corporation Systems and methods for diagnosing faults in a multiple domain storage system
US7882393B2 (en) * 2007-03-28 2011-02-01 International Business Machines Corporation In-band problem log data collection between a host system and a storage system
US7987383B1 (en) * 2007-04-27 2011-07-26 Netapp, Inc. System and method for rapid indentification of coredump disks during simultaneous take over
US20080270638A1 (en) * 2007-04-30 2008-10-30 International Business Machines Corporation Systems and methods for monitoring high speed network traffic via simultaneously multiplexed data streams
US7936767B2 (en) * 2007-04-30 2011-05-03 International Business Machines Corporation Systems and methods for monitoring high speed network traffic via sequentially multiplexed data streams
US7779308B2 (en) * 2007-06-21 2010-08-17 International Business Machines Corporation Error processing across multiple initiator network
KR101421054B1 (en) * 2007-08-06 2014-07-18 삼성전자주식회사 Method and system for distributing operation by using buffer
US7913106B2 (en) * 2007-12-18 2011-03-22 International Business Machines Corporation Failover in a host concurrently supporting multiple virtual IP addresses across multiple adapters
US8866698B2 (en) * 2008-10-01 2014-10-21 Pleiades Publishing Ltd. Multi-display handheld device and supporting system
US20110258377A1 (en) * 2009-12-07 2011-10-20 Hitachi, Ltd. Disk array system and command processing method for disk array system
US20120044050A1 (en) * 2010-08-23 2012-02-23 Samir Vig Smart Doorbell Security System and Method to Identify Visitors
US9027019B2 (en) * 2011-09-22 2015-05-05 Cisco Technology, Inc. Storage drive virtualization
JP5959733B2 (en) * 2013-04-23 2016-08-02 株式会社日立製作所 Storage system and storage system failure management method
WO2015011749A1 (en) * 2013-07-22 2015-01-29 株式会社日立製作所 Storage system and fault management method for storage system
EP3271820B1 (en) * 2015-09-24 2020-06-24 Hewlett-Packard Enterprise Development LP Failure indication in shared memory

Family Cites Families (51)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4200930A (en) * 1977-05-23 1980-04-29 Burroughs Corporation Adapter cluster module for data communications subsystem
DE69032508T2 (en) * 1989-12-22 1999-03-25 Tandem Computers Inc Fault-tolerant computer system with online reinsert and shutdown / start
US5287519A (en) * 1992-09-17 1994-02-15 International Business Machines Corp. LAN station personal computer system with controlled data access for normal and unauthorized users and method
US5901327A (en) * 1996-05-28 1999-05-04 Emc Corporation Bundling of write data from channel commands in a command chain for transmission over a data link between data storage systems for remote data mirroring
JPH086854A (en) * 1993-12-23 1996-01-12 Unisys Corp Outboard-file-cache external processing complex
US6353599B1 (en) * 1995-10-16 2002-03-05 Nec Corporation Wireless enumeration
US5727218A (en) * 1996-03-05 1998-03-10 Unisys Corp. Controlling an apparatus disposed for adapting fiber channel transmissions to an industry standard data bus
US5890207A (en) * 1996-11-27 1999-03-30 Emc Corporation High performance integrated cached storage device
US5987621A (en) * 1997-04-25 1999-11-16 Emc Corporation Hardware and software failover services for a file server
US6226680B1 (en) * 1997-10-14 2001-05-01 Alacritech, Inc. Intelligent network interface system method for protocol processing
US6219693B1 (en) * 1997-11-04 2001-04-17 Adaptec, Inc. File array storage architecture having file system distributed across a data processing platform
US6009478A (en) * 1997-11-04 1999-12-28 Adaptec, Inc. File array communications interface for communicating between a host computer and an adapter
US6009463A (en) * 1998-04-15 1999-12-28 Unisys Corporation Cooperative service interface with buffer and lock pool sharing, for enhancing message-dialog transfer between network provider and distributed system services
US6101533A (en) * 1998-04-15 2000-08-08 Unisys Corporation Multiple interface data communication system and method utilizing multiple connection library interfaces with buffer and lock pool sharing
JP4392877B2 (en) 1998-09-18 2010-01-06 株式会社日立製作所 Disk array controller
US6594700B1 (en) * 1999-06-14 2003-07-15 International Business Machines Corporation System and method for implementing a universal service broker interchange mechanism
US6665714B1 (en) * 1999-06-30 2003-12-16 Emc Corporation Method and apparatus for determining an identity of a network device
US6480901B1 (en) * 1999-07-09 2002-11-12 Lsi Logic Corporation System for monitoring and managing devices on a network from a management station via a proxy server that provides protocol converter
US6625750B1 (en) * 1999-11-16 2003-09-23 Emc Corporation Hardware and software failover services for a file server
US6658504B1 (en) * 2000-05-16 2003-12-02 Eurologic Systems Storage apparatus
US6795862B1 (en) * 2000-05-31 2004-09-21 International Business Machines Corporation System for converting a version of SNMP entered by user into another version used by device and providing default values for attributes not being specified
JP2002014878A (en) 2000-06-30 2002-01-18 Toshiba Corp Computer system and its maintenance method
US6807581B1 (en) * 2000-09-29 2004-10-19 Alacritech, Inc. Intelligent network storage interface system
US7143249B2 (en) * 2000-10-04 2006-11-28 Network Appliance, Inc. Resynchronization of mirrored storage devices
AU2002246717A1 (en) * 2000-12-21 2002-08-19 Imtech Corporation Method and apparatus for displaying information on a large scale display
US7260636B2 (en) * 2000-12-22 2007-08-21 Emc Corporation Method and apparatus for preventing unauthorized access by a network device
US7165096B2 (en) * 2000-12-22 2007-01-16 Data Plow, Inc. Storage area network file system
US20020087720A1 (en) * 2000-12-28 2002-07-04 Davis Arlin R. System and method for communications management and control over an unreliable communications network
WO2002065298A1 (en) * 2001-02-13 2002-08-22 Candera, Inc. Silicon-based storage virtualization server
US6606690B2 (en) * 2001-02-20 2003-08-12 Hewlett-Packard Development Company, L.P. System and method for accessing a storage area network as network attached storage
US20020129216A1 (en) * 2001-03-06 2002-09-12 Kevin Collins Apparatus and method for configuring available storage capacity on a network as a logical device
US6779063B2 (en) * 2001-04-09 2004-08-17 Hitachi, Ltd. Direct access storage system having plural interfaces which permit receipt of block and file I/O requests
US20020161982A1 (en) * 2001-04-30 2002-10-31 Erik Riedel System and method for implementing a storage area network system protocol
US6944133B2 (en) * 2001-05-01 2005-09-13 Ge Financial Assurance Holdings, Inc. System and method for providing access to resources using a fabric switch
US20030037061A1 (en) * 2001-05-08 2003-02-20 Gautham Sastri Data storage system for a multi-client network and method of managing such system
JP4632574B2 (en) * 2001-05-25 2011-02-23 株式会社日立製作所 Storage device, file data backup method, and file data copy method
JP2003015826A (en) * 2001-07-04 2003-01-17 Hitachi Ltd Shared memory copy function in disk array controller
JP4156817B2 (en) * 2001-07-27 2008-09-24 株式会社日立製作所 Storage system
JP2003162439A (en) * 2001-11-22 2003-06-06 Hitachi Ltd Storage system and control method therefor
JP3964212B2 (en) * 2002-01-16 2007-08-22 株式会社日立製作所 Storage system
JP2003316713A (en) * 2002-04-26 2003-11-07 Hitachi Ltd Storage device system
JP2003345631A (en) * 2002-05-28 2003-12-05 Hitachi Ltd Computer system and allocating method for storage area
US7493404B2 (en) * 2002-05-30 2009-02-17 Lsi Corporation Apparatus and method for providing transparent sharing of channel resources by multiple host machines utilizing mixed mode block and file protocols
US7020770B2 (en) * 2002-07-12 2006-03-28 Lsi Logic Corporation Method and apparatus for configuration of RAID controllers
JP4175083B2 (en) * 2002-10-29 2008-11-05 株式会社日立製作所 Storage device management computer and program
JP4252301B2 (en) * 2002-12-26 2009-04-08 株式会社日立製作所 Storage system and data backup method thereof
JP2004220216A (en) * 2003-01-14 2004-08-05 Hitachi Ltd San/nas integrated storage device
JP4322031B2 (en) * 2003-03-27 2009-08-26 株式会社日立製作所 Storage device
JP2005108098A (en) * 2003-10-01 2005-04-21 Hitachi Ltd Data i/o device and control method of data i/o device
JP2005267111A (en) * 2004-03-17 2005-09-29 Hitachi Ltd Storage control system and method for controlling storage control system
US7275175B2 (en) * 2004-07-22 2007-09-25 International Business Machines Corporation Method and apparatus for high-speed network adapter failover

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9063853B2 (en) 2011-05-31 2015-06-23 Fujitsu Limited Storage device, storage system, and method for controlling storage device
EP2854020A1 (en) 2013-09-27 2015-04-01 Fujitsu Limited Storage control apparatus, storage control method, and storage control program
US9483211B2 (en) 2013-09-27 2016-11-01 Fujitsu Limited Storage control apparatus, storage control method, and computer-readable recording medium having stored storage control program
WO2016020979A1 (en) * 2014-08-05 2016-02-11 株式会社日立製作所 Computer system and intermediary device
WO2016075765A1 (en) * 2014-11-12 2016-05-19 株式会社日立製作所 Computer system and control method therefor
JPWO2016075765A1 (en) * 2014-11-12 2017-06-29 株式会社日立製作所 Computer system and control method thereof
US10353613B2 (en) 2014-11-12 2019-07-16 Hitachi, Ltd. Computer system and control method therefor for handling path failure
US9928185B2 (en) 2015-01-30 2018-03-27 Fujitsu Limited Information processing apparatus and computer-readable recording medium having program recorded therein
JP2018524751A (en) * 2015-07-22 2018-08-30 華為技術有限公司Huawei Technologies Co.,Ltd. Computer device and method for reading / writing data by computer device

Also Published As

Publication number Publication date
US7697312B2 (en) 2010-04-13
EP1439453A2 (en) 2004-07-21
EP1439453A3 (en) 2008-07-02
US20040139168A1 (en) 2004-07-15
US7185143B2 (en) 2007-02-27
US20070168559A1 (en) 2007-07-19

Similar Documents

Publication Publication Date Title
US9298379B2 (en) Bi-directional data transfer within a single I/O operation
US20170286243A1 (en) Failover Handling In Modular Switched Fabric For Data Storage Systems
US9052829B2 (en) Methods and structure for improved I/O shipping in a clustered storage system
US8607230B2 (en) Virtual computer system and migration method of virtual computer
USRE47289E1 (en) Server system and operation method thereof
US10402363B2 (en) Multi-port interposer architectures in data storage systems
US8074002B2 (en) Route aware serial advanced technology attachment (SATA) switch
TWI439867B (en) Dynamic physical and virtual multipath i/o
US7523235B2 (en) Serial Advanced Technology Attachment (SATA) switch
US6170067B1 (en) System for automatically reporting a system failure in a server
JP4252301B2 (en) Storage system and data backup method thereof
US7617360B2 (en) Disk array apparatus and method of controlling the same by a disk array controller having a plurality of processor cores
US6341322B1 (en) Method for interfacing two buses
JP4477906B2 (en) Storage system
KR101252903B1 (en) Allocation-unit-based virtual formatting methods and devices employing allocation-unit-based virtual formatting methods
US8312189B2 (en) Processing of data to monitor input/output operations
DE10297278B4 (en) Network storage device for connecting a host computer to at least one memory device and method for zoning a memory controller module (CMM) to a channel interface module (CIM) within a network storage device (100)
US7519851B2 (en) Apparatus for replicating volumes between heterogenous storage systems
US7181578B1 (en) Method and apparatus for efficient scalable storage management
US7664909B2 (en) Method and apparatus for a shared I/O serial ATA controller
US7028177B2 (en) Array controller ROM cloning in redundant controllers
US5784617A (en) Resource-capability-based method and system for handling service processor requests
JP4383148B2 (en) Magnetic disk array device with processing offload function module
US6970972B2 (en) High-availability disk control device and failure processing method thereof and high-availability disk subsystem
US6961813B2 (en) System and method for providing multi-initiator capability to an ATA drive

Legal Events

Date Code Title Description
A621 Written request for application examination

Free format text: JAPANESE INTERMEDIATE CODE: A621

Effective date: 20050926

RD01 Notification of change of attorney

Free format text: JAPANESE INTERMEDIATE CODE: A7421

Effective date: 20060420

A977 Report on retrieval

Free format text: JAPANESE INTERMEDIATE CODE: A971007

Effective date: 20080801

A131 Notification of reasons for refusal

Free format text: JAPANESE INTERMEDIATE CODE: A131

Effective date: 20080826

A521 Written amendment

Free format text: JAPANESE INTERMEDIATE CODE: A523

Effective date: 20081010

A131 Notification of reasons for refusal

Free format text: JAPANESE INTERMEDIATE CODE: A131

Effective date: 20090728

A521 Written amendment

Free format text: JAPANESE INTERMEDIATE CODE: A523

Effective date: 20090831

A02 Decision of refusal

Free format text: JAPANESE INTERMEDIATE CODE: A02

Effective date: 20100223